World-wide industrial experience shows that vibration is a powerful tool in detecting damage in mechanical systems which made it very widely used in rotating and reciprocating machines. Properly used vibration monitoring systems provide information illustrating what is on-going in the machine, i.e. if there is deterioration. Current and past data can also be used to identify what type of deterioration processes are active. But, sometimes the vibration signal analysis should be completed by additional CM parameter(s), such as SPM, oil analysis and temperature, to get more reliable detection of problems and prediction of future condition of production equipment.
Making accurate and cost-effective maintenance decisions, i.e. to determine more accurately when and why to stop production equipment and where to conduct maintenance actions depends on:
- availability of equipment history,
- possibility to assess current condition,
- accuracy of the CM system being used and
- the knowledge and experience in monitoring the condition of mechanical systems that is available.
But, the data set which describes the condition of the equipment in the close future is usually missed. Therefore to avoid failures and conduct maintenance actions during planned stoppages, it is crucial to assess what the condition of production equipment will be at the next planned stoppage (close future), i.e. when the decisions of stopping production equipment for maintenance should be realised.
The capital invested in production equipment is usually intended to produce a predetermined amount of production in a well-defined production time at a particular quality and production cost. But due to failures, reduced production speed and other unnecessary stoppages, the production process loses some of the time planned for production due to the downtime and failure related poor quality. Higher level of vibration, noise, heat generation and temperature can associate the deterioration phase preceding failures and result in higher energy consumption. The additional energy, i.e. waste in energy, is consumed due to, e.g. failures, reduced production speed, produced bad quality items and machine idle time. Thus, eliminating the root-causes behind these disturbances, the bleeding of waste in energy consumption can be reduced or terminated. Moreover, the pollution released to the environment is, in general, influenced by many factors, among others, energy production/consumption. Therefore, reduced waste in energy consumption will reduce the amount of pollution released to the environment. It is notable that the reduction in downtime of production equipment and waste in energy consumption can both be assessed on the economic basis but of two dimensional effect; economic and environmental. Therefore, in order to decide whether the time is most profitable, technical and economic analysis is important for highlighting the economic losses due to failures, downtimes and unnecessary consumption of energy and consequently less pollution to the environment. It is also important to develop a model for determining when to plan maintenance actions to continuously increase company profitability. The economic impact of unnecessary downtime and consumption in energy can then be used to distinguish between failure modes, rank them and select the most profitable maintenance time. But, applying any model or system for conducting profitable maintenance cannot be secured without reliable and easily used key performance indicators (KPIs) that can be used for mapping, analysis, evaluation and judgement of maintenance and production processes. A KPI is a measurable variable that can be used for mapping, analysis or evaluating a process technically, e.g. with respect to failures, downtime, short stoppages, or economically, with respect to production cost, economic losses, maintenance savings. It reflects either only technical measure, such as the number of failures, or economic measure, such as the economic losses due to inefficient maintenance, or a combined measure, such as maintenance savings per high quality item.
The detailed analysis of CM signals is essential for detecting problems at an early stage, identifying root-causes, predicting damage severity and, planning and selecting the most profitable maintenance time and actions. The criteria influencing the arrangements and selection of CM systems and maintenance management policy will be controlled by important features;
1. diagnosis accuracy,
2. cost-effectiveness of maintenance decisions,
3. applicability of the system in SMEs from technical, administrative and economic perspectives.
4. Level of knowledge and experience in CM technology that is required for adequate implementation.
From everyday experience, more accurate and high quality maintenance and operational data that are necessary for describing the strengths and weaknesses of production equipment are also important for modifying the design of production equipment. Also, maintenance role in modifying equipment design through providing the underlying information and evidences that are required to highlight weaknesses and justify modifications technically and economically.
One of the biggest problems facing the implementation of CBM and especially predictive maintenance is that reliable predictive maintenance demands according to the current way a big number of sensors and may be of different types, e.g. for measuring vibration, SPM, acoustic emission, etc. At continuous monitoring of the health of production equipment, the problem becomes bigger due to the high investing capital that is demanded in many cases.